Traffic: Why We Drive the Way We Do (and What It Says About Us)

By Tom Vanderbilt

2008, 416 pages

Price on Amazon.com: US$ 16.47

‘Traffic: Why We Drive the Way We Do (and What It Says About Us)’ is directed just as much at the average driver as it is towards the urban or transportation planner. It is written in a very lively manner to attract the interest of someone who otherwise would not buy a book on the topic of traffic. It deals very much with the psychology of driving as well as the mechanics of driving. By understanding more about the minds of drivers, we can plan for a city that takes understands the limitations of the driver, so that cars can co-exist with pedestrians more and so that streets may be safer. This is important because we drive worse than we think we do; the New York Times book review even suggests another title for the book might be ‘Idiots.’

In some ways this book can be compared one of the great books on urban studies: City by William Whyte (although not as innovative as City). Both books try to explain to the reader the human element. City shows how the human actually uses the space in a park or in a subway stop. Many urban designers design a place how they want it to be used. While in ‘Traffic’, it is the human and driving. It is not about how we would like the human to drive, but how that person actually does drive. Many traffic engineers design a road how they want the driver to use the road. The problem is that the human may actually use a space or drive much differently than we want them to, and we need to take that into account when designing spaces and roads.

Some of the book is common knowledge for planners, such as:

  1. building more roads won’t relieve congestion in the long-term,
  2. the anonymity of driving can help explain traffic rage.
  3. human behavior affects traffic flow on a highway.

But most of the book is filled with information that will make you re-think driving. Much of the beginning has parts on traffic flow, technology to track driver behavior and other technologies. As you read through it there are more and more lines in the book you want to underline, either for the often counter-intuitive (反直觉)concepts  or for the interesting facts.

One of his better chapters is titled ‘When Dangerous Roads are Safer’, which is about how we, the human, change our behavior in response to perceived risk (181). It is hard hard to design roads to be absolutely safe because ‘safety features meant to reduce the consequences of driver error encourage drivers to drive in a way that requiring those generous safety features.’ (209)The road itself tells us much more than signs do, which may be why adding reflector posts to a curved road resulted in higher speeds and more accidents than when there were no posts. The road seemed safer, so people drove more dangerously. Traffic engineers have been trying for ‘design consistency’, to let drivers know what to expect and then give it to them. The problem is that too much expectancy can be boring, and boring makes people not pay attention. Interchanges seem the most dangerous part of a highway, but most people lose their lives on straight stretches of ‘designed-for-safety’ highways when a single vehicle has run off the road (because the driver has drifted off to sleep).

The lesson from one part of that chapter is: ‘when a situation feels dangerous to you, its probably more safe than you know; when a situation feels safe, that is exactly when you should be careful. Remember, most crashes happen on dry roads, on clear, sunny days to sober (没喝醉的;不过量饮酒的) drivers.’ (185)

And it can apply just as equally to pedestrians too. Pedestrians relax and aren’t as careful looking for cars when the pedestrian has the green light to cross the street. The result? More urban pedestrians are killed while legally crossing the street than while jaywalking (jaywalk: 不守交通规则横穿马路). Jaywalkers are careful when they cross the street because they know they are doing something dangerous, and thus it may actually become less dangerous. The person legally crossing the street assumes all cars will be obeying traffic rules, and thus sometimes pays less attention

In Chapter eight he discusses the various driving and pedestrian cultures around the world, such as why New Yorkers love to jaywalk but people in Copenhagen don’t. As in chapters on the history of traffic in the west he touches the history of driving culture, such as why the British and the Japanese drive on the left side, but other countries drive on the right. In arguing why pedestrians seem to disobey traffic rules in some countries more than other countries the well-known urban planner from Denmark, Jan Gehl, said, ‘The more you make things difficult for pedestrians, the more you downgrade their status in the traffic system, then the more they will take the law into their own hands.’ (225).

The number of traffic fatalities per capita rise with per capita GDP up to a certain point, and then start to drop. So fatalities per capita will generally vary with GDP. But sometimes countries that have similar levels of GDP have different levels of traffic risk. The example used in the book is between the Netherlands and Belgium, two countries that are alike in many ways, but in Belgium the fatal and non-fatal crash rate is almost twice as high as in the Netherlands. The explaining factor is corruption. (Belgium is 9th least corrupt country while Belgium is 20th.) Even though the traffic laws are similar in both countries, corruption can explain different attitudes to following and enforcing the laws. Corruption seems to reduce the legitimacy of the authorities in the eyes of the people, and makes people more resistant to following laws. And in enforcing, the Netherlands issued eight times the number of traffic tickets than did Belgium. The statistical link between corruption and traffic fatalities is stronger than with income and traffic fatalities.

The final chapter discusses why we think driving is safer than it is, and risk factors for unsafe driving. Driving seems safe because driving fatalities on a per-trip basis is so small, even though, on a lifetime basis, the numbers seem more dangerous.[1] Driving is voluntary, in our control, and there is a reward (getting to the destination). And each safe trip we take just reinforces in our mind that driving is safe. Our perception of our driving ability and the safety of driving is based on the number of crashes we’ve been in, rather than on the number of accidents we have avoided (and the average person has just narrowly avoided a serious crash more often than they think they have). But all you need is one bad crash.

Just like making roads safer can encourage risky behavior, the safer cars get, the more risk drivers choose to take. The explanation seems to be in something called ‘risk homeostasis’. This theory implies that people have a ‘target level’ of risk. So as something becomes safer, people make up for it by doing something else riskier. Overall risk level ends up being the same.

Also, people change their behavior more strongly when there is direct feedback. That is why people in small cars take less risk than those in larger cars. The driver in a small car is closer to the ground, can feel more road noise, and feels the sensation of moving fast on a highway, and so recognizes the potential danger, unlike a driver of an SUV, who is more insulated from the road. During snowstorms people drive more cautiously because they can directly feel less control over the vehicle. So, although there are more accidents in snowstorms, there are fewer deaths because people have been driving slower.

In China many of the ways to lessen traffic fatalities is to lessen conflicts between automobile and pedestrian/bicyclist. And the most popular way to lessen conflicts seems to be restricting the pedestrian, rather than the vehicle. This sometimes creates more friction than it reduces. The best quote in the book, is by the great English author of the 1800’s, Charles Dickens. He said, ‘Most people would rather face the danger of the street, rather than the fatigue of getting upstairs.’ This is a man speaking 150 years ago, but the quote is very relevant to the difficulty of trying to get pedestrians to use pedestrian bridges.

Also relevant to China is the authors remarks on curb radii.  The intersections of newer parts of cities often have curb corners with large radii, which encourages the driver to take the turn quickly, while doing nothing to remind the driver about the pedestrian that may be crossing legally.

However, some of the design methods in Europe that have reduced traffic accidents the most have been those that give priority to the pedestrian. Sometimes the car needs to be shown that, in many parts of the city, it is the guest in social world, rather than the social world being a guest in the world of the vehicle.

This is a very good book that will be of interest to the general urban planner or designer, and also get the transportation engineer to think more about something that the transportation engineer can’t control-us, the person.

And remember, we should not underestimate the stupidity of us as drivers. Drivers looking at crashes often get into crashes themselves-and we people do like to look at crashes (163). Think about that next time you see a crash. Better yet, just keep your eyes on the road.


[1]For example, “For every 100 million miles that are driven in vehicles in the United States, there are 1.3 deaths.” That sentence seems less dangerous than, “If you drive an average of 15,500 miles per year, as many Americans do, there is about a 1 in 100 chance you will die in a car crash over a lifetime of 50 years of driving”

Interesting facts from the book:

  • The bigger intersections grow, the less efficient they become (162).
  • Stop signs do little to reduce speed-drivers just go faster at the midblock location to make up time (190)
  • Unlike developed countries, most people dying in traffic in developing countries are dying not in cars, but outside. In Delhi, India, only 5% of traffic fatalities are occupants of cars, but pedestrians, cyclists and motorcyclists make up 80%. (233)
  • With the huge increase of new drivers in Beijing, the insurance industry has had to adjust, with some insurers reporting accident risk for certain classes of individuals at almost 100%–meaning that that class of driver was not an ‘accident risk’ but an ‘accident certainty.’ (233)
  • The more densely populated a place, the higher the non-fatal crash rate. The less densely populated a place, the higher the traffic fatality rate.
  • In one study of 5,000 car-truck collisions, in 70% of the cases the driver of the car had the sole contributing responsibility for the crash (247). So, even though people fear driving near trucks, they should really fear their own behavior around trucks.
  • Except for teenage drivers, people drive safer when there is a passenger in the car.
  • The fatality rate in the backseat is 26% lower than in the front. Backseats are safer than airbags. (263)
  • Up to 80% of crashes occur from people not paying attention for as little as 3 seconds.
  • Most traffic jams are ‘non-recurring congestion’, meaning that the road normally functions fine and is perhaps congested maybe due to construction or weather, but most often gets congested due to crashes. So the solution for easing congestion there is for people to get into fewer crashes and not build more highway lanes (163).

The first of the book’s nine chapters is available for free at the New York Times website.

There is also a traffic quiz on questions related to the book there.

(I also have the book if anyone would like to borrow it. But someone borrowed City from me about six years ago and never returned it.)

Advertisements